Hi all,
I remember reading an article in the IEEE Communications magazine awhile back about a study that was done on the effect of an SCBA mask on radio communications. I think I tossed the magazine, but I looked it up on their website and it was the Jan 2006 edition if anyone cares to look it up.
I seem to remember it talking about the effect of SCBA voice distortion and background noise on the vocoder in digital radio systems. I'm wondering what kind of effect a P.A.S.S. device going off in the background would have on a P25 vocoder. The algorithms in vocoders are based on the fact the human speech is fairly predictable and has certian known characteristics. I'm wondering how the vocoder manages when a very loud "non-human speech" noise (like a PASS alarm going off) is present at the mic in addition to the desired speech. I think the article covered some of this sort of thing and the data looked a little disconcerting...I'm wishing now I hadn't thrown that magazine away during a cleaning binge.
I would think that this was thought about when the vocoder was developed, but I'd be interested in hearing from anyone who has experience with this. I'd hate to have a firefighter not be able to get help because the radio's vocoder gets confused. You could have 100% signal strength but still get garbage.
Any ideas?
Thanks in advance,
Andy
P25 Vocoder in Firefighting Situations
Moderator: Queue Moderator
-
High_order1
- Posts: 210
- Joined: Sun Oct 13, 2002 10:39 am
Hey Andy
I have access to the IEEE library and I have the article in the HD. If you want I can PM it to you.
From what I've read about P25 systems, IMBE is heavily optimized on human speech, so it won't do a good job on handling non-human voice. Given this advantage, any unwanted background noise will be easily suppressed in the IMBE encoding and transmitted.
On the other hand in the case of desired signal (like the PASS device), it would probably just sound like a garbled digital burst on the receiver end. Keep in mind that IMBE is for human speech, not for machine generated sound (in this case, an emergency device that most likely will emit a high frequency alarm), from what I understand, the frequency response of most P25 vocoder is centered on the lower edge and won't handle high frequency audio nicely.
Feel free to correct whatever I've said wrong, I'm just a student!
I have access to the IEEE library and I have the article in the HD. If you want I can PM it to you.
From what I've read about P25 systems, IMBE is heavily optimized on human speech, so it won't do a good job on handling non-human voice. Given this advantage, any unwanted background noise will be easily suppressed in the IMBE encoding and transmitted.
On the other hand in the case of desired signal (like the PASS device), it would probably just sound like a garbled digital burst on the receiver end. Keep in mind that IMBE is for human speech, not for machine generated sound (in this case, an emergency device that most likely will emit a high frequency alarm), from what I understand, the frequency response of most P25 vocoder is centered on the lower edge and won't handle high frequency audio nicely.
Feel free to correct whatever I've said wrong, I'm just a student!
-
Tom in D.C.
- Posts: 3859
- Joined: Tue Sep 04, 2001 4:00 pm
- What radios do you own?: Progreso soup can with CRT
Fireground comms...
Maybe this is one reason why NFPA has recommended that analog be
used on the fireground. In my area this recommendation is
completely ignored wherever a P25 system is now in use.
used on the fireground. In my area this recommendation is
completely ignored wherever a P25 system is now in use.
Tom in D.C.
In 1920, the U.S. Post Office Department ruled
that children may not be sent by parcel post.
In 1920, the U.S. Post Office Department ruled
that children may not be sent by parcel post.
an anecdote
Well, I cannot speak to what happens when you use a SCBA system, or have an alarm going off, but I can attest to this:
I was playing around with the IMBE vocoder and our code that feeds a test voice sample into it. For fun, I replaced the standard intelligibility test sample ("We took the crosstown bus...") with the opening 4 lines of Kansas's "Carry on Wayward Son" - which are just a Capella voice harmony.
The vocoder freaked. The resulting output, while recognizable as "Carry On My Wayward Son..." was really funky sounding - like it was being spoken through a fan. In fact, at first I was cross-checking the code to make sure I didn't break something.
It was just that the vocoder decided that all these harmonically related voices must be once voice being chopped.
The problem is that there is a great deal of difference between a single voice, with no background noise (which is what the bulk of the voice clarity tests are) and a voice with sirens, alarms, hissing air tanks, crackling fire, whooshing water, etc.
I was playing around with the IMBE vocoder and our code that feeds a test voice sample into it. For fun, I replaced the standard intelligibility test sample ("We took the crosstown bus...") with the opening 4 lines of Kansas's "Carry on Wayward Son" - which are just a Capella voice harmony.
The vocoder freaked. The resulting output, while recognizable as "Carry On My Wayward Son..." was really funky sounding - like it was being spoken through a fan. In fact, at first I was cross-checking the code to make sure I didn't break something.
It was just that the vocoder decided that all these harmonically related voices must be once voice being chopped.
The problem is that there is a great deal of difference between a single voice, with no background noise (which is what the bulk of the voice clarity tests are) and a voice with sirens, alarms, hissing air tanks, crackling fire, whooshing water, etc.
Last edited by Wowbagger on Wed Oct 18, 2006 11:24 am, edited 1 time in total.
This is my opinion, not Aeroflex's.
I WILL NOT give you proprietary information. I make too much money to jeopardize my job.
I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
I WILL NOT give you proprietary information. I make too much money to jeopardize my job.
I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
I got the article from the IEEE library at work today, but thanks for the offer dxon2m!
Basically the study they did was on the effect that an SCBA mask has on speech intelligibility. One of the more interesting tests they did was to have people speak standardized sentences and then have someone else listen and see how many words they got wrong.
As expected, the SCBA mask made it harder to understand the speaker. What’s more troubling, though, is that when they ran the SCBA-distorted audio through the P25 codec, the word error rate went up even more, and that was without background noise of any kind. The rough numbers are something like a 2% word error rate with no mask analog, 3% no mask with codec, 5% with mask analog, and between 7 and 13% with mask and codec.
Now what happens if you throw in a bunch of background noise? If the codec reproduces the background noise accurately, no big deal, as that’s the same situation as analog. But we know that doesn’t happen. If the codec pukes and can’t figure out what to do with all of the noise, nothing gets through but garble. I think Wowbagger’s test drives this point home.
And maybe just to clarify, I’m not so much concerned about the PASS alarm noise making it through the radio…if the codec suppresses it, great. My concern is with a firefighter trying to talk on his radio while his PASS is activated. I want his voice to be audible, regardless of whether you hear the PASS going off in the background or not. If the PASS noise messes with the vocoder and makes the voice unreadable, that’s a bad situation.
So my next follow up question becomes…is the FCC effectively forcing us into digital because of its current obsession with spectrum efficiency (which isn’t really a bad thing) and narrowbanding? Will the push for spectrum efficient communications hurt intelligibility in the long run?
I think the NFPA’s recommendation makes a lot of sense, but will it be legal in years to come?
Andy
Basically the study they did was on the effect that an SCBA mask has on speech intelligibility. One of the more interesting tests they did was to have people speak standardized sentences and then have someone else listen and see how many words they got wrong.
As expected, the SCBA mask made it harder to understand the speaker. What’s more troubling, though, is that when they ran the SCBA-distorted audio through the P25 codec, the word error rate went up even more, and that was without background noise of any kind. The rough numbers are something like a 2% word error rate with no mask analog, 3% no mask with codec, 5% with mask analog, and between 7 and 13% with mask and codec.
Now what happens if you throw in a bunch of background noise? If the codec reproduces the background noise accurately, no big deal, as that’s the same situation as analog. But we know that doesn’t happen. If the codec pukes and can’t figure out what to do with all of the noise, nothing gets through but garble. I think Wowbagger’s test drives this point home.
And maybe just to clarify, I’m not so much concerned about the PASS alarm noise making it through the radio…if the codec suppresses it, great. My concern is with a firefighter trying to talk on his radio while his PASS is activated. I want his voice to be audible, regardless of whether you hear the PASS going off in the background or not. If the PASS noise messes with the vocoder and makes the voice unreadable, that’s a bad situation.
So my next follow up question becomes…is the FCC effectively forcing us into digital because of its current obsession with spectrum efficiency (which isn’t really a bad thing) and narrowbanding? Will the push for spectrum efficient communications hurt intelligibility in the long run?
I think the NFPA’s recommendation makes a lot of sense, but will it be legal in years to come?
Andy
Wholly apart from speech intelligibility, I believe that, on the basis of an incident that took place on FDNY a couple of years where simplex portables employing digital voice encoding and located close to one another failed to synch (and thus nothing came through), NFPA has opined that digital voice is not acceptable on the fireground.
There can also be issues of how vocoders handle non-voice messages, particularly those involving pure tones (such as alerts, evacs, and the like).
There can also be issues of how vocoders handle non-voice messages, particularly those involving pure tones (such as alerts, evacs, and the like).
-
MTS2000des
- Posts: 3347
- Joined: Sat Jan 04, 2003 4:59 pm
- What radios do you own?: XTS2500, XTS5000, and MTS2000
pretty obvious to me that P25 phase I really isn't ready for primetime in all public safety enviornments, especially fireground hot zones. I guess the engineers didn't think of these variables like SCBA and how important "background audio" can be in a PS enviornment. We don't have this problem with good ol Armstrong FM.
The views here are my own and do not represent those of anyone else or the company, the boss, his wife, his dog or distant relatives.
I was really hoping someone would say "yeah they thought of that a long time ago and designed the P25 standard taking that into consideration". I don't want to sound all "anti digital" because I'm not...there are a lot of great things you can do with digital that you can't do with analog. It just has to be designed with the application in mind from the get go. As for not being able to pass evac tones, etc. through a digital radio, that's true in a sense, but if the system is designed properly you should be able to send a digital evac signal that would cause the radio to emit a tone. Tone signalling is an analog signalling format...when you switch to digital voice you also switch to digital signalling.
Andy
Andy
-
fogster
- Posts: 386
- Joined: Sun Nov 06, 2005 10:38 am
- What radios do you own?: XTS2500/5000, XPR7550/5550
Re: an anecdote
Have you ever thought about posting some of the resulting clips? It might be interesting, and certainly a good asset. (Of course, I don't know if it'd be permitted by your employer?)Wowbagger wrote:The vocoder freaked. The resulting output, while recognizable as "Carry On My Wayward Son..." was really funky sounding - like it was being spoken through a fan.
I'd love to see the FCC try explaining to firefighters that spectral efficiency is more important than their safety. At the least, they should allow a few 'legacy' analog, 'wideband' FM channels to exist...I think the NFPA’s recommendation makes a lot of sense, but will it be legal in years to come?
-
MTS2000des
- Posts: 3347
- Joined: Sat Jan 04, 2003 4:59 pm
- What radios do you own?: XTS2500, XTS5000, and MTS2000
that's the ironic part, P25 phase I isn't any more "spectrally efficient" than narrowband FM. Phase I is still 1:1 subscriber unit to channel ratio. Phase II (which isn't out yet) uses TDMA and offers more spectral efficiency than Phase I or analog FM. Problem is, it is ANOTHER transition in itself to go from Phase I to Phase II.
Guess who gets to sell more radios and systems! What a scam and sham this is.
P25 Phase I was merely an answer to a problem that didn't exist. Back in the early 90's when Motorola and EGE rolled out APCO 16 compliant digital air interfaces (read Astro VSELP and EGE Aegis) there was paranoia about "incompatible digital modulation". BFD. Most of the systems are proprietary closed network trunked systems (like Smartnet and in the case of EGE Provoice) so who cares?
Just another piece of junk technology that isn't 10 times better than what it replaced (NBFM) it's just 10 times more expensive and now that everyone's got it, we are seeing the artifacts (like the lack of real world performance in hot zone fireground applications) where a pair of FRS radios have more value than 5,000 dollar digital P25 Phase I radios do.
Can someone tell me what real world advantage P25 Phase I has over NBFM? Other than "it's digital"? I mean, you can get all or most of the touted features with a signalling system like Stat-Alert. Oh sure, a digital signal is more immune to interference in some cases...well it just masks it or poor radio system design.(Maybe if we didn't let some bloated ESMR impersonate a celphone company on interleaved channels we wouldn't have such a high noise floor on 800MHz to begin with.) But is Phase I really worth the hassle? Why didn't we hold out to Phase II or (down the road) Phase III? Oh I know, cause we have to make Ed Zander's stock go up "a quarter of a point".
Guess who gets to sell more radios and systems! What a scam and sham this is.
P25 Phase I was merely an answer to a problem that didn't exist. Back in the early 90's when Motorola and EGE rolled out APCO 16 compliant digital air interfaces (read Astro VSELP and EGE Aegis) there was paranoia about "incompatible digital modulation". BFD. Most of the systems are proprietary closed network trunked systems (like Smartnet and in the case of EGE Provoice) so who cares?
Just another piece of junk technology that isn't 10 times better than what it replaced (NBFM) it's just 10 times more expensive and now that everyone's got it, we are seeing the artifacts (like the lack of real world performance in hot zone fireground applications) where a pair of FRS radios have more value than 5,000 dollar digital P25 Phase I radios do.
Can someone tell me what real world advantage P25 Phase I has over NBFM? Other than "it's digital"? I mean, you can get all or most of the touted features with a signalling system like Stat-Alert. Oh sure, a digital signal is more immune to interference in some cases...well it just masks it or poor radio system design.(Maybe if we didn't let some bloated ESMR impersonate a celphone company on interleaved channels we wouldn't have such a high noise floor on 800MHz to begin with.) But is Phase I really worth the hassle? Why didn't we hold out to Phase II or (down the road) Phase III? Oh I know, cause we have to make Ed Zander's stock go up "a quarter of a point".
The views here are my own and do not represent those of anyone else or the company, the boss, his wife, his dog or distant relatives.
First of all, let me say (takes off Aeroflex badge) that in general I agree with the statements about digital vs. analog. However, there are a few advantages to digital vs. analog:
1) Repeaters do not add any more distortion to the signal, as they regenerate it - so a marginal signal into the repeater (as long as the BER is less than the ability of the forward error correction) can be repeated without any change.
2) Encryption. The analog voice scrambling systems are trivial to defeat, and add a degree of distortion. Once you are digital, encryption does not add any distortion, and current computational quantum mechanics says you'd need a computer the size of a small star (and as hot) to brute-force 256 AES.
3) Access control. No matter what, you are going to have to have a digital data stream somewhere on the channel to do access control (e.g. trunking). You can either run a low speed low frequency data stream under the voice (Passport, LTR, analog EDACS, Smartnet) or you can take the whole thing digital.
However, I think you ALL are overlooking the real driving force behind the govt. wanting everybody to standardize on APCO-25, which has almost NOTHING to do with the voice channel per se.
The idea behind APCO-25 is that is FULLY SPECIFIES both the voice channel and the control channel. The problem with systems like SmartNet is that the control channel protocol is proprietary, so that all your radio vendors HAVE to have licensed that standard from the vendor who created it. That's why the government is trying to transition everybody from Smartnet w/ APCO voice channels (3600 baud Smartnet control channel, 4800 baud C4FM voice channel) to full APCO-25 trunking systems (4800 baud C4FM or CQPSK control channel, 4800 baud traffic channel). The fact that the voice channel is digital is just future-proofing.
(Gets out soapbox).
However, what I don't like is how the TIA keeps picking proprietary vocoders like IMBE and AMBE - as a result, to make an APCO-25 compatible radio, you *have* to license the vocoder from DVSI. In my own experimentation, I have found that Speex will do as well compressing voice to the same bit rates that AMBE will, and is freely (in both the Beer and Speech senses) available. I've talked to the guys at work who go to the TIA committee meetings and asked them to try to bring this up, but I am not holding my breath.
This is also why I don't like the D-STAR and AOR Digital HF modes - again, they are *NOT* "open" as they are claimed precisely because they are based around AMBE. Everything else about those modes excites me - I've been lamenting for years that, just because the FCC rules state "2400 baud max" for 2 meters does not mean 2400 bits per second - BAUD != BPS. You can get 9600 BPS on 2400 baud if you run 16QAM, 16FM, or 16PSK (or OFDM, or ....) If people had actually THOUGHT about it, packet could have been at least reasonably close to dial-up speeds.
(Puts away the soapbox).
1) Repeaters do not add any more distortion to the signal, as they regenerate it - so a marginal signal into the repeater (as long as the BER is less than the ability of the forward error correction) can be repeated without any change.
2) Encryption. The analog voice scrambling systems are trivial to defeat, and add a degree of distortion. Once you are digital, encryption does not add any distortion, and current computational quantum mechanics says you'd need a computer the size of a small star (and as hot) to brute-force 256 AES.
3) Access control. No matter what, you are going to have to have a digital data stream somewhere on the channel to do access control (e.g. trunking). You can either run a low speed low frequency data stream under the voice (Passport, LTR, analog EDACS, Smartnet) or you can take the whole thing digital.
However, I think you ALL are overlooking the real driving force behind the govt. wanting everybody to standardize on APCO-25, which has almost NOTHING to do with the voice channel per se.
The idea behind APCO-25 is that is FULLY SPECIFIES both the voice channel and the control channel. The problem with systems like SmartNet is that the control channel protocol is proprietary, so that all your radio vendors HAVE to have licensed that standard from the vendor who created it. That's why the government is trying to transition everybody from Smartnet w/ APCO voice channels (3600 baud Smartnet control channel, 4800 baud C4FM voice channel) to full APCO-25 trunking systems (4800 baud C4FM or CQPSK control channel, 4800 baud traffic channel). The fact that the voice channel is digital is just future-proofing.
(Gets out soapbox).
However, what I don't like is how the TIA keeps picking proprietary vocoders like IMBE and AMBE - as a result, to make an APCO-25 compatible radio, you *have* to license the vocoder from DVSI. In my own experimentation, I have found that Speex will do as well compressing voice to the same bit rates that AMBE will, and is freely (in both the Beer and Speech senses) available. I've talked to the guys at work who go to the TIA committee meetings and asked them to try to bring this up, but I am not holding my breath.
This is also why I don't like the D-STAR and AOR Digital HF modes - again, they are *NOT* "open" as they are claimed precisely because they are based around AMBE. Everything else about those modes excites me - I've been lamenting for years that, just because the FCC rules state "2400 baud max" for 2 meters does not mean 2400 bits per second - BAUD != BPS. You can get 9600 BPS on 2400 baud if you run 16QAM, 16FM, or 16PSK (or OFDM, or ....) If people had actually THOUGHT about it, packet could have been at least reasonably close to dial-up speeds.
(Puts away the soapbox).
This is my opinion, not Aeroflex's.
I WILL NOT give you proprietary information. I make too much money to jeopardize my job.
I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
I WILL NOT give you proprietary information. I make too much money to jeopardize my job.
I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
Re: Fireground comms...
I'm not aware of NFPA having any such recommendation. NFPA 1221, Standard for Emergency Service Communication only requires a separate channel from the radio dispatch channel for on-scene tactical comminication. The standard permits analog as well as digital. There are additional requirements for digital systems where they are required to revert to conventional mode if there was a failure in the system.Tom in D.C. wrote:Maybe this is one reason why NFPA has recommended that analog be
used on the fireground. In my area this recommendation is
completely ignored wherever a P25 system is now in use.
Re: an anecdote
OK, you asked, you get.fogster wrote:Have you ever thought about posting some of the resulting clips? It might be interesting, and certainly a good asset. (Of course, I don't know if it'd be permitted by your employer?)Wowbagger wrote:The vocoder freaked. The resulting output, while recognizable as "Carry On My Wayward Son..." was really funky sounding - like it was being spoken through a fan.
Test conditions:
2975, being fed the audio via the microphone input in dynamic mike mode. Audio provided by my Nokia 770, playing back Kansas's "Carry On Wayward Son", the opening bars, under Fair use of the US copyright code.
Audio output from the 2975 being routed into a computer and captured as a WAV file via the computer's sound card.
The first part of the file is the audio going through as narrowband FM and being received as narrowband IF (12.5 kHz IF bandwidth). The second part is going through as APCO-25 conventional mode, C4FM.
A further test (not in this file) was feeding in a single voice in the same manner - in that case the voice was not as distorted going through the APCO-25 path.
The wav is NOT compressed, nor was it properly anti-aliased by the soundcard - so you will hear some aliasing noise on the analog FM path.
This is my opinion, not Aeroflex's.
I WILL NOT give you proprietary information. I make too much money to jeopardize my job.
I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
I WILL NOT give you proprietary information. I make too much money to jeopardize my job.
I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
-
MTS2000des
- Posts: 3347
- Joined: Sat Jan 04, 2003 4:59 pm
- What radios do you own?: XTS2500, XTS5000, and MTS2000
Wowbagger, no arguement that a full digital stream has alot more capability in the area of encryption and access control. But think about it, not everyone needs encyrption and access control itself hinders interoperability (which is the touted catchphrase). The question I have is why the rush to implement digital radio systems now that obviously are still in their infancy realitve to "legacy analog" which has been pretty much at it's zenith as far as performance, reliability and cost effectiveness. P25 Phase I is just in itself a stepping stone to Phase II which promises some real world avantages over Phase I. Why not wait for a more solid, tested digital platform than the "version 1" of digital voice before we all start relying on it for mission critical applications? Like any software, it takes years of development to overcome major bugs. People's lives are too much the price to pay to find out this stuff doesn't work.
The views here are my own and do not represent those of anyone else or the company, the boss, his wife, his dog or distant relatives.
-
bellersley
- No Longer Registered
- Posts: 872
- Joined: Tue Feb 22, 2005 7:03 am
I know Shaun (r0f) posted some really good P25 test files back in the day. Maybe someone should repost a bunch of various tests. I'll volunteer to do so if nobody else wants to.
I'm thinking maybe Analog, Analog Securenet, Astro, Astro Securenet. Then record pure clean voice with each, then voice with background noise, then voice with alarms going off (like a P.A.S.S) then finally music, to see what happens.
If I can find the time, is anyone able to host these WAV files for me?
I'm thinking maybe Analog, Analog Securenet, Astro, Astro Securenet. Then record pure clean voice with each, then voice with background noise, then voice with alarms going off (like a P.A.S.S) then finally music, to see what happens.
If I can find the time, is anyone able to host these WAV files for me?
I read a lot here, but post very little.
As a line firefighter for 16 years, occasionally on the sharp end, a few things are missing from the argument. One thing about digital is that the signal is either there or not, the analog signal dissolves down into the squelch/hash before dropping out. A good dispatcher or fire alarm operator can sometimes pick out a word or several syllables as it's fading down into the squelch. With digital you wouldn't have the opportunity to try and decipher what's not there. (?!)
Also, the other half of the equation is that the fire departments, especially in suburbia and the smaller cities are poor stepchildren to law enforcement. If it's a standalone system for the fire department usually it won't be funded fully.
For me, the bottom line is that for on scene fireground operations, it really needs to be simplex analog. Most of these new systems are so over engineered it's ridiculous. Chiefs get 'wowed' by new technology and usually want a bigger, better and more feature laden system than the Chief next door. The manufacturers are more than happy to add new gadgets, regardless of whether it will benefit the firefighters or the citizens.
As a line firefighter for 16 years, occasionally on the sharp end, a few things are missing from the argument. One thing about digital is that the signal is either there or not, the analog signal dissolves down into the squelch/hash before dropping out. A good dispatcher or fire alarm operator can sometimes pick out a word or several syllables as it's fading down into the squelch. With digital you wouldn't have the opportunity to try and decipher what's not there. (?!)
Also, the other half of the equation is that the fire departments, especially in suburbia and the smaller cities are poor stepchildren to law enforcement. If it's a standalone system for the fire department usually it won't be funded fully.
For me, the bottom line is that for on scene fireground operations, it really needs to be simplex analog. Most of these new systems are so over engineered it's ridiculous. Chiefs get 'wowed' by new technology and usually want a bigger, better and more feature laden system than the Chief next door. The manufacturers are more than happy to add new gadgets, regardless of whether it will benefit the firefighters or the citizens.
Listen to the man.
All of our wonderful radios have actually no purpose whatsoever except as a tool to be used by the grunt in the field in order to help him do his job more effectively, more quickly, and with less risk to him or those he is trying to help. This, above all else, should be the acid test of any "improvement" any of us geniuses come up with.
All of our wonderful radios have actually no purpose whatsoever except as a tool to be used by the grunt in the field in order to help him do his job more effectively, more quickly, and with less risk to him or those he is trying to help. This, above all else, should be the acid test of any "improvement" any of us geniuses come up with.
-
MTS2000des
- Posts: 3347
- Joined: Sat Jan 04, 2003 4:59 pm
- What radios do you own?: XTS2500, XTS5000, and MTS2000
the problem is in many cases people who lack knowledge of technology are responsible for making the decisions to implement such new systems. Only after it fails miserably are questions asked. In many cases, the responsible decision makers only have the data that the vendor provides and it is nearly impossible for them to get factual, objective information from qualified persons regarding the new technology.RKG wrote:Listen to the man.
All of our wonderful radios have actually no purpose whatsoever except as a tool to be used by the grunt in the field in order to help him do his job more effectively, more quickly, and with less risk to him or those he is trying to help. This, above all else, should be the acid test of any "improvement" any of us geniuses come up with.
The way that life safety critical radio systems are designed, sold and implemented need to change. We don't always need "cutting edge" technology. What we do need is field proven, reliable, SIMPLE and AFFORDABLE solutions. As radio amateurs, we have more RELIABLE and SIMPLE communications solutions at our fingertips. We know that it doesn't take trunking radios, infrastructure and digital modulation for one person to talk to another in the same building. Adding anything else to this equation further increases the points of failure.
there is NO room for points of failure on a fireground or hot zone situation anywhere for that matter.
"Keep It Simple Stupid"
The views here are my own and do not represent those of anyone else or the company, the boss, his wife, his dog or distant relatives.
To quote Putt's law:MTS2000des wrote:
the problem is in many cases people who lack knowledge of technology are responsible for making the decisions to implement such new systems. Only after it fails miserably are questions asked. In many cases, the responsible decision makers only have the data that the vendor provides and it is nearly impossible for them to get factual, objective information from qualified persons regarding the new technology.
"Technology is dominated by two types of people: those who understand what they do not manage, and those who manage what they do not understand." - Archibald Putt
The book is a must read...absolutely hilarious but oh so true. It's published by the IEEE's Wiley press.
http://www.wiley.com/WileyCDA/WileyTitl ... 14224.html
Andy
UPDATE
It looks like the IAFC has found out about this issue and is trying to collect data from the field:
http://www.iafc.org/displaycommon.cfm?a ... clenbr=719
Andy
http://www.iafc.org/displaycommon.cfm?a ... clenbr=719
Andy
Well, technically it's not digital modulation that's the problem, but the parametric vocoder that assumes a certain human speech pattern. If the vocoder isn't designed to handle background noise, the results will be unpredictable. It's entirely possible to sample an analog waveform and convert it to digital without using a vocoder...just send the raw ones and zeros over the air without any modification. That would result in a perfectly reproduced signal on the receiving end (assuming a decent sampling rate and number of A/D bits). However, doing this is not spectrally efficient. The vocoder is what allows the spectral efficiency. A good analogy would be a .wav file versus a .mp3 file. A .wav file is uncompressed...it contains all of the original information. A .mp3 file is compressed...an algorithm determines what information can be "thrown out" to save space while still retaining enough information for the signal to be reproduced acceptably. The same thing is true of a real-time vocoder. The encoding algorithm is essentially "compressing" the sampled signal in order to reduce the amount of information being sent and therefore reduce the amount of spectrum needed to send the signal.
As I've said before...when lives are at risk, spectral efficiency be damned...robustness should come first, spectral efficiency second.
You guys are doing some very valuable testing in Boise...I'm glad you're making it known. Keep up the good work!
Andy
As I've said before...when lives are at risk, spectral efficiency be damned...robustness should come first, spectral efficiency second.
You guys are doing some very valuable testing in Boise...I'm glad you're making it known. Keep up the good work!
Andy
I've done some testing between a pair of XTS's I have here (DSP N08.03.03 and N08.03.02....bfd...not a big diff). Clean voice is spectacular as you would expect a new DSP to sound. However, with noise in the backround (turned up the radio on 95.1 will rock/chicago) and it sounded like someone was mumbling in the backround, hence a very "pixelated" bacround noise as a result.
-
Nexrad16
- Batboard $upporter
- Posts: 336
- Joined: Sun Mar 21, 2004 8:01 pm
- What radios do you own?: APX V/7 & V/U Liberty
I must admit that I'm not that educated in the digital vs vocoder as you appear to be. That might be part of the trouble in that the radio equipment maker and the ff's don't speak the same language. We want clear communication = 2 or better than our vhf ana. The radio maker just wants to... Well anyway it's a huge deal. We have a huge amount of taxpayer money tied up in this unacceptable system that is worse than what we have now (we will not transition to 700 unless this issue is resolved).
We do favor 700 but not in its present state and not with a bunch of gadgets, wires or simplex.
We continue to monitor this site for info that might be of value to our current issue. Our intent is to provide for the safety of our ff's, other public safety professions & the public.
If anyone has expertise to the issue please contact the IAFC or the IAFF. If you wish to contact me please pm me with your personal email address and I will provide you with my city email address (@cityofboise.org).
Thanks everyone!
We do favor 700 but not in its present state and not with a bunch of gadgets, wires or simplex.
We continue to monitor this site for info that might be of value to our current issue. Our intent is to provide for the safety of our ff's, other public safety professions & the public.
If anyone has expertise to the issue please contact the IAFC or the IAFF. If you wish to contact me please pm me with your personal email address and I will provide you with my city email address (@cityofboise.org).
Thanks everyone!
You don't want to be relying on any sort of trunking system inside a burning building, digital or otherwise.Nexrad16 wrote:We do favor 700 but not in its present state and not with a bunch of gadgets, wires or simplex.
What happens if you really need to call for help, but your radio just goes bah-bah-bah-bah... "SYSTEM BUSY"?
As it stands, 700 digital being good for wide area dispatch operations doesn't automatically mean it's good for local incident communications.
A digital radio system suitable for fireground, vertical rescue, confined space operations and whatnot is not beyond the realm of possibility - the intrateam radios you see developed for military applications are proof enough of that.
http://www.rfcomm.harris.com/spr/
(Four users talking simultaneously in full duplex, anyone?)
Unfortunately, nobody in the public safety radio market have pulled their heads in and taken a look at what sort of communications you need at somewhere like a fire. Hopefully nobody needs to get killed while trying to get out on a 700MHz digital trunking system from a portable in a burning basement for this to be looked at as a serious issue.
-
Nexrad16
- Batboard $upporter
- Posts: 336
- Joined: Sun Mar 21, 2004 8:01 pm
- What radios do you own?: APX V/7 & V/U Liberty
I'll greatly limit my response on the debate about simplex vs trunked on the fg only to say that if that is the case then we just spent a huge amount of local, state , fed tax dollars for a system that works no better than what we had. In fact we have less interop as the BLM is on vhf.
Many agencies use their trunk sys on the fg. Salt Lake UT, Dayton OH, Huber Heights OH, and some 20 other agencies with in Montgomery County OH. All with success. Although analog those agencies have no plan for using simplex as normal fg ops. Enought said, my part is to gather info and attempt to find a cure, simplex is not an option as normal fg ops.
Simplex digital is poor at best during our testing.
I have been part of analog 800 trunked and commo is very clear with vibra alert, saws, fans etc.. That sys was /\/\ using MTS2000's
Again thanks for the input.
Many agencies use their trunk sys on the fg. Salt Lake UT, Dayton OH, Huber Heights OH, and some 20 other agencies with in Montgomery County OH. All with success. Although analog those agencies have no plan for using simplex as normal fg ops. Enought said, my part is to gather info and attempt to find a cure, simplex is not an option as normal fg ops.
Simplex digital is poor at best during our testing.
I have been part of analog 800 trunked and commo is very clear with vibra alert, saws, fans etc.. That sys was /\/\ using MTS2000's
Again thanks for the input.
-
MTS2000des
- Posts: 3347
- Joined: Sat Jan 04, 2003 4:59 pm
- What radios do you own?: XTS2500, XTS5000, and MTS2000
you said it "we just spent a huge amount on a system that works no better than what we had" and this is the understatement of the century. This is a prime example of what is wrong with the whole spec, bid and buy process for public safety communications. The vendors, (Motorola especially) have way too much influence and control.Nexrad16 wrote:I'll greatly limit my response on the debate about simplex vs trunked on the fg only to say that if that is the case then we just spent a huge amount of local, state , fed tax dollars for a system that works no better than what we had. In fact we have less interop as the BLM is on vhf.
Many agencies use their trunk sys on the fg. Salt Lake UT, Dayton OH, Huber Heights OH, and some 20 other agencies with in Montgomery County OH. All with success. Although analog those agencies have no plan for using simplex as normal fg ops. Enought said, my part is to gather info and attempt to find a cure, simplex is not an option as normal fg ops.
Simplex digital is poor at best during our testing.
I have been part of analog 800 trunked and commo is very clear with vibra alert, saws, fans etc.. That sys was /\/\ using MTS2000's
Again thanks for the input.
The buyers often lack non-partisan real factual data to make informed purchases and most rely on what their service provider or radio shop tells them they need. Groups that are supposed to be independent and represent public safety interests are often funded by large donations from those very radio vendors. Common sense tells you that someone who is trying to sell you something isn't going to reveal all the imperfections of their wares. It is up to the buyer to do his own research but how when he doesn't even know what he is looking at or who to trust?
One of the great resources in your community are radio amateurs, many of which actually have professional lives in wireless communications, but when we have our "ham hats" on, most of us think reliable, simple and practical not to mention economical when it comes to solving problems. Case in point, cross patching radios. We've been doing this long before P25, fancy audio bridges, even existed. In the 80's we had field programmable dual band radios (such as the Kenwood TM-721) that would allow us to patch two radios, repeaters or whatever from V to U with ease.
KEEP IT SIMPLE is a rule that absolutely needs to be followed when it comes to MISSION CRITICAL LIFE SAFETY communications. Relying on infrastructure in a fire ground hot zone is an accident waiting to happen, and the risk management department at any government would ban the use of trunking system talkgroups for fiire grounds if they actually understood how these systems actually work and how many possible failure points there are between radio A and radio B. The average ham knows if you're within simplex range, then move off the repeater to simplex. The points of failure are limited to the individual users and/or their radios. There are no repeaters, remote receivers, controllers, audio paths or links to fail. Coverage is usually not an issue unless you are dealing with multi-story buildings, in which case crews can relay transmissions manually. That's still better than the "out of range" tone where NO ONE can hear you AT ALL.
As I said before, digital modulation with compression removes a portion of audio spectrum that is just as valuable as the voice traffic. As a dispatcher, I get a mental picture of what the officer is dealing with not just by his voice or inflection, but nuances in the background (people yelling, sounds of traffic, sirens, etc) and my personal experience with P25 is those nuances are often distorted or not present at all. The folks who designed the vocoder are thinking efficiency and there is no reason to be efficient when it's people's lives that are at risk. I say it's a flawed technology and we should not allow it to become "tombstone" technology like the commercial aviation industry often does before we decide it's broken and needs fixing.
Or we could just stay with FM which has been pretty much perfected and has been field proven until such time as digital is just as clean sounding and predictable as FM is and not risk people's lives playing guinea pigs. Let's let us hams do the playing....
The views here are my own and do not represent those of anyone else or the company, the boss, his wife, his dog or distant relatives.
The police and ambulance here are about to switch from analog conventional to digital conventional. I fully expect massive failure and a return to analog within two weeks of the switch. Actually, now that I think about it. I fully expect massive failure but I wouldn't put it pasr them to continue using it despite the massive failure.
"I'll eat you like a plate of bacon and eggs in the morning. "
- Some loser on rr.com
eBay at it's finest:
Me: "What exactly is a 900Mhz UHF CB?"
Them: "A very nice CB at 900Mhz speed!"
- Some loser on rr.com
eBay at it's finest:
Me: "What exactly is a 900Mhz UHF CB?"
Them: "A very nice CB at 900Mhz speed!"
Our fire department uses APCO 25 radios that were purchased with "grant money". They are programmed with analog simplex channels, which is what should be used at the fire scene. Even knowing that analog simplex will work better, there are problems with using it.
The problem is, there are far too many things happening for me to wade through a bunch of menus to find the simplex channel. True, this could be addressed through how the radio is programmed, but often times all the radios in the county have to have the same template, which means that no department will have their talkgrounds and simplex channels on the same knob twist.
If I have to look at a display or find a menu button, I have had to pay too much attention to the radio. Everything has to be able to be done by feel, just like I do with the air tank controls, nozzle, etc. The best that can be hoped for is that I twist the channel knob CCW until it stops and then count clicks.
This all seems obvious to me, and to many others reading, but apparently some people don't understand it because they keep buying us $5000 toys that are of no use when crawling around in smoke and heat.
How about a 2 channel MT500?
The problem is, there are far too many things happening for me to wade through a bunch of menus to find the simplex channel. True, this could be addressed through how the radio is programmed, but often times all the radios in the county have to have the same template, which means that no department will have their talkgrounds and simplex channels on the same knob twist.
If I have to look at a display or find a menu button, I have had to pay too much attention to the radio. Everything has to be able to be done by feel, just like I do with the air tank controls, nozzle, etc. The best that can be hoped for is that I twist the channel knob CCW until it stops and then count clicks.
This all seems obvious to me, and to many others reading, but apparently some people don't understand it because they keep buying us $5000 toys that are of no use when crawling around in smoke and heat.
How about a 2 channel MT500?
I programmed all of our radios with the fireground simplex as channel 1 and channel 16 in both groups. That way you if you get in trouble, you can bang the knob all the way in either direction and you are where you need to be. No futzing around and no need to ever even look at the radio when you pick it up.
"I'll eat you like a plate of bacon and eggs in the morning. "
- Some loser on rr.com
eBay at it's finest:
Me: "What exactly is a 900Mhz UHF CB?"
Them: "A very nice CB at 900Mhz speed!"
- Some loser on rr.com
eBay at it's finest:
Me: "What exactly is a 900Mhz UHF CB?"
Them: "A very nice CB at 900Mhz speed!"
Our Kenwoods have two buttons on the speaker mic... black takes you to dispatch, orange takes you to fireground. The guys love it - no thinking required.nmfire10 wrote:I programmed all of our radios with the fireground simplex as channel 1 and channel 16 in both groups. That way you if you get in trouble, you can bang the knob all the way in either direction and you are where you need to be. No futzing around and no need to ever even look at the radio when you pick it up.
-
High_order1
- Posts: 210
- Joined: Sun Oct 13, 2002 10:39 am
I monitor my state's progress in the Great Switch To The Promised Band, but I tend to lurk and not talk.
I thought I would ask about what you guys are discussing, because it seems that this problem is much, much greater than just fireground.
What about a medic calling base to relay vitals in the back of a truck in heavy traffic and running code?
What about a Squad gal at an accident trying to request assistance with the spreader and half a dozen people screaming?
What about a SWAT Assaulter taking rounds and trying to keep an element from coming around into an exposed areas?
All these scenarios seem like plausible circumstances where the vocoder's firmware would throw up its' hands and mute the radio.
The response from them:
Is there any more data on this currently I can provide this guy besides the Phoenix tests and the current IAFF survey that backs this story up?
Thanks,
-Shawn
I thought I would ask about what you guys are discussing, because it seems that this problem is much, much greater than just fireground.
What about a medic calling base to relay vitals in the back of a truck in heavy traffic and running code?
What about a Squad gal at an accident trying to request assistance with the spreader and half a dozen people screaming?
What about a SWAT Assaulter taking rounds and trying to keep an element from coming around into an exposed areas?
All these scenarios seem like plausible circumstances where the vocoder's firmware would throw up its' hands and mute the radio.
The response from them:
To my knowledge there is no problem. We use P25 digital radio in high noise environment as well as Nashville Metro PD, FD, EMS, EMA and many other public safety agencies.
Memphis FD has been digital for even longer than Nashville as well as FD's all over the US.
Is there any more data on this currently I can provide this guy besides the Phoenix tests and the current IAFF survey that backs this story up?
Thanks,
-Shawn
-
nickburns186
- Posts: 48
- Joined: Sat Feb 19, 2005 8:40 pm
Gonna give my two cents here......About 2 years ago when a fire department I do work for switched to digital conventional, they had read some articles and heard the stories so they decided to do their own tests, and I was present. They went through the whole gamut of PASS devices sounding, talking wearing an air mask, sirens sounding, etc....and to be honest, they didnt have a bit of trouble with the audio being distorted any more than analog. They were pleased. The firmware of the radios tested, about 40 of them, have been kept up to date over the years and I'm not hearing any complaints. These are 5000's BTW.
That seems to be a problem where I live, too. What's worse is that the customers can get brainwashed by the salesmen because there aren't enough neutral folks around to talk some sense into them and the salesmen are waving shiny new toys in front of their faces.MTS2000des wrote:The buyers often lack non-partisan real factual data to make informed purchases and most rely on what their service provider or radio shop tells them they need. Groups that are supposed to be independent and represent public safety interests are often funded by large donations from those very radio vendors. Common sense tells you that someone who is trying to sell you something isn't going to reveal all the imperfections of their wares. It is up to the buyer to do his own research but how when he doesn't even know what he is looking at or who to trust?
) take care of the issue at hand, then use the radio when the smoke clears 
http://mrtmag.com/mobile_voice/news/iaf ... ue-032307/
MRT wrote:IAFC seeks input regarding noise issue with digital radios
Mar 23, 2007 11:03 AM, By Donny Jackson
Common fireground noise, such as some alerting systems and active power tools, can render voice communications from digital radios unintelligible, the International Association of Fire Chiefs (IAFC) said this week in an alert issued to its members.
The IAFC has received reports that voice communications from firefighters using digital radios cannot be heard when the radios are operated in close proximity to the low-pressure alarm of their self-contained breathing apparatus (SCBA), according to Charles Werner, IAFC member and fire chief of Charlottesville, Va.
Upon learning of the issue, Werner had members of his department conduct tests, one of which tested whether a digital-radio voice call could be understood if a chainsaw was being operated three feet away.
“What we found is that, when that loud noise was going off, the voice became totally unintelligible,” Werner said. “When we did the test, you couldn’t tell that the person even said a word.
“In analog mode, you can hear the chainsaw loud in the background, but you can distinguish clearly the person talking. So, we narrowed it down to the fact that it’s a digital problem.”
Specifically, the problem appears to be a problem with the vocoder’s used in public-safety digital radios, regardless of the manufacturer of the device or the frequency band used, Werner said.
“Some of the technical people are saying that the vocoders doing exactly what it supposed to be doing—taking all the noise away and letting the voice be intelligible,” he said. “But the noise is so loud that it’s overwhelming the voice, which is being lost.
“What surprised me was that these digital radios that have been sold to fire departments across the country were never tested in these environments.”
With this in mind, Werner said he will become a member of a working group established by IAFC that will work with vendors in an effort to address these problems. Meanwhile, the IAFC has asked its members to submit any findings they have experienced in such circumstances during tests or in the field.
“The important thing is that we’re hoping we prevent everyone from thinking that digital radio is bad,” he said. “We’ve gained a great deal of benefits and features from the digital side—improvements in coverage, capacity, interoperability and emergency-alerting features—we just have to figure out how we work our way through this to fix this anomaly.”
Very interesting reading. While my Fire Department is still in the Analog world, we know we will eventually have to go "Digital" at some point. I have been following confersations on this subject for a while, and have not seen any reference to "Digital-Feedback". My fire Department uses external radio speakers on just about every device they own. When they respond to an event, "feed-back" is a common occurance, and they are completely able to talk over / through the feed-back with little affect on the audio's intellengence. Enter "Digital-feedback". With just a few tests, we have determined that this may be our hardest problem to solve. Has anyone tried to talk over a digital channel with another radio active? Completely impossible.
Larry
Larry